HyperNova 60B 2605 vs Step 3.7 Flash
Which is better for the money: HyperNova 60B 2605 by Multiverse Computing or Step 3.7 Flash by StepFun? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-08-07.
Trade-off
Verdict
Depends on your budget
Step 3.7 Flash scores 12.6 points higher (30.9 vs 18.3) but costs 8.5× as much to run the benchmark ($320 vs $37.5; mid-range vs budget cost classes). They are not alternatives at one budget: pick Step 3.7 Flash when you need intelligence above 18.3, and HyperNova 60B 2605 when 18.3 is enough.
| Metric | HyperNova 60B 2605 | Step 3.7 Flash | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 18.3 | 30.9 | +68.9% Step 3.7 Flash is 68.9% more intelligent |
| Total AA cost (lower is better) | $37.50 | $319.97 | +753.3% Step 3.7 Flash is 8.5× more expensive |
| End-to-end time (lower is better) | 6.93s | 7.79s | +12.4% Step 3.7 Flash is 12.4% slower |
| Price per 1M tokens (in / out) | $0.04 / $0.14 | $0.20 / $1.15 | |
| Input modalities | Not reported | Not reported |
Step 3.7 Flash vs HyperNova 60B 2605: 68.9% more intelligence, 753.3% more cost to run the AA benchmark, and 12.4% slower end-to-end.
Today's Pareto frontier
Against the frontier
- HyperNova 60B 2605 is off the frontier: at the same or lower cost, GPT-5.6 Luna (medium) scores 38.90 for $21.1.
- Step 3.7 Flash is off the frontier: at the same or lower cost, GPT-5.6 Luna (max) scores 52.30 for $172.
HyperNova 60B 2605Step 3.7 FlashToday's Pareto frontierEvery other model
Matchup
Compare something else
Agents:
Accept: text/markdown or ?format=md returns this page as Markdown. JSON: /api/v1/compare/hypernova-60b/step-3-7-flash · Markdown: /api/v1/compare/hypernova-60b/step-3-7-flash?format=md · llms.txt · agent skill